85 lines
2.6 KiB
Python
85 lines
2.6 KiB
Python
from flask import Flask, request, jsonify
|
|
import os
|
|
import uuid
|
|
import subprocess
|
|
from werkzeug.utils import secure_filename
|
|
import time
|
|
import random
|
|
|
|
app = Flask(__name__)
|
|
UPLOAD_FOLDER = '/data'
|
|
ALLOWED_EXTENSIONS = {'wav', 'mp3', 'ogg', 'flac'}
|
|
|
|
def allowed_file(filename):
|
|
return '.' in filename and filename.rsplit('.', 1)[1].lower() in ALLOWED_EXTENSIONS
|
|
|
|
def generate_filename():
|
|
timestamp = int(time.time())
|
|
random_num = random.randint(1000, 9999)
|
|
return f"{timestamp}-{random_num}"
|
|
|
|
@app.route('/stt', methods=['POST'])
|
|
def transcribe():
|
|
if 'audio' not in request.files:
|
|
return jsonify({"error": "No audio file provided"}), 400
|
|
|
|
file = request.files['audio']
|
|
if file.filename == '':
|
|
return jsonify({"error": "Empty filename"}), 400
|
|
|
|
if not file or not allowed_file(file.filename):
|
|
return jsonify({"error": "Invalid file type"}), 400
|
|
|
|
try:
|
|
# Generate filename with timestamp + random number
|
|
file_base = generate_filename()
|
|
orig_ext = secure_filename(file.filename).rsplit('.', 1)[1].lower()
|
|
orig_path = os.path.join(UPLOAD_FOLDER, f"{file_base}.{orig_ext}")
|
|
wav_path = os.path.join(UPLOAD_FOLDER, f"{file_base}.wav")
|
|
output_base = os.path.join(UPLOAD_FOLDER, file_base)
|
|
|
|
# Save original file
|
|
file.save(orig_path)
|
|
|
|
# Convert to WAV if needed
|
|
if orig_ext != 'wav':
|
|
subprocess.run([
|
|
'ffmpeg', '-i', orig_path,
|
|
'-ar', '16000', '-ac', '1', '-c:a', 'pcm_s16le', wav_path
|
|
], check=True)
|
|
os.remove(orig_path)
|
|
audio_path = wav_path
|
|
else:
|
|
audio_path = orig_path
|
|
|
|
# Run whisper.cpp
|
|
subprocess.run([
|
|
'/app/whisper.cpp/build/bin/main',
|
|
'-m', '/app/whisper.cpp/models/ggml-tiny.en-q5_1.bin',
|
|
'-f', audio_path,
|
|
'-otxt', '-of', output_base
|
|
], check=True)
|
|
|
|
# Read output
|
|
output_file = f"{output_base}.txt"
|
|
with open(output_file, 'r') as f:
|
|
transcription = f.read()
|
|
|
|
# Return the filenames in the response
|
|
response = {
|
|
"text": transcription.strip(),
|
|
"audio_file": f"{file_base}.{orig_ext}",
|
|
"text_file": f"{file_base}.txt"
|
|
}
|
|
|
|
return jsonify(response)
|
|
|
|
except subprocess.CalledProcessError as e:
|
|
return jsonify({"error": f"Whisper processing failed: {e.stderr}"}), 500
|
|
except Exception as e:
|
|
return jsonify({"error": str(e)}), 500
|
|
|
|
if __name__ == '__main__':
|
|
os.makedirs(UPLOAD_FOLDER, exist_ok=True)
|
|
app.run(host='0.0.0.0', port=4004)
|